The second main issue is the effect of a trade ban on ivory harvest. Illegal ivory constituted 80% of all ivory traded prior to the CITES ban (ITRG, 1989). It is, therefore, important to understand the determinants of poaching (Milner-Gulland & Leader-Williams, 1992), and especially how poaching is influenced by international trade. A simple model is presented below to demonstrate the interaction between legal and illegal supplies in the international ivory market. The model highlights some unintended side effects of the trade ban, which tend to be neglected in the conservation debate.2

Thus, the main effects of a trade ban on legal and illegal markets are as follows. One, the moral impact of a trade ban (and associated attention surrounding the plight of the African elephant) is to reduce demand, which leads to lower poaching offtake. The second effect is the withdrawal of official stocks (from culling operations etc) from the markets. This draws up the black market price of ivory and induces more poaching. Third, if a trade ban facilitates additional interceptions of smuggled ivory, this has an ambiguous effect on poaching incentives. Interceptions reduce smugglers' expected price and, hence, the price they are willing to pay poachers for ivory, reducing poaching. However, under a trade ban confiscations cannot be marketed.

The conclusion is neither that markets for products from threatened natural resources should be liberalised indiscriminately nor that strict trade bans should form the core of conservation efforts. Rather, it should be investigated how legal marketing channels for official stocks can be set up and safeguarded, as in the current system with limited ivory export quotas from a group of Southern African range states. Due attention should be paid to efforts aimed at developing secure techmques to distinguish legal from illegal supplies, i.e. through a product marking system.

Whatever effect these changes had on how men organized themselves socially and politically in relation to the hunt, it and the related activities of caravan trading and porterage had a distinct effect simply through the number of men they drew out of the pool of labour available for work in the community (Alpers, 1992, p. 356). Trade, which caused this problem, also supplied its solution: more slave labour purchased with the wealth generated by trade. This labour was not only applied to subsistence and domestic main- tenance left: 133.5

Arab traders returning from the interior brought back tales of great riches in ivory to be had almost for the taking. European travellers added to and embroidered these stories. Cameron, who journeyed across Africa in 1874, met Arabs 'bound for lands of fabulous riches...where ivory was reported to be used for fencing pig-styes and making door posts'.7 Livingstone relates that in Manyemaland, through which he travelled in I872,

The two great inland markets for ivory were Unyanyembe (Tabora) in what is now central Tanzania, and Ujiji on the east coast of Lake Tanganyika.10 From Tabora routes branched to the north, to Uganda, to the west, and to the south and Lake Rukwa. At Unyanyembe and Ujiji, Arab merchants had set themselves up in style, surrounding themselves with the coconut palms of their Zanzibar home, and living in cool tembes, waited on by slaves, and comforted by concubines-reproducing the languid environment of the spice island. At Unyanyembe the Arab merchant from Zanzibar met his compatriot returning from the Lake or Karagwe, and here much bartering and trade took place. Porters hired on the shores of Lake Targanyika were paid off, and a fresh gang collected from those discharged y a previous caravan arriving from the coast, which in turn would take cn the porters laid off by a down-going caravan